B3108 Speed Limit Analysis

Introduction

The B3108 runs north of Winsley conurbation connecting Warminster Road in the West, with Bath Road in Bradford on Avon.

There have been on-going discussions regarding the speed limit on this road. This document aims to expand on some of the points mentioned in Briefing Note: “B3108 Winsley Bypass” published in November 2022.

• That speeds are already on average in the same ballpark as proposed speed limits
  
• Cost implications of slower speeds on journey times are balanced out by other costs, such as collision costs, noise and benefits from adding infrastructure to enable car free trips. TAG guidance reflect more value than time for an employed person using the road.
• That a significant number of people live along the road, with numerous properties fronting onto the road.

Typical speeds

In order to determine the typical speeds of vehicles travelling along the road the Google Distance API was used. This is a tool to estimate the journey time between two points for a future date based on historic data. Using the “best guess” parameters provides the most realistic single prediction available from the model by integrating real-time road conditions with historical patterns https://developers.google.com/maps/documentation/distance-matrix/distance-matrix. It returns the estimated time to travel this distance based on historic data. This has been compared against measured data and found to be a good representation of vehicle speed.

The road was split into 6 sections, based on the Open Street Map (OSM) segments, with some segments joined together to simplify the model and reduce the number of calls to the API. Described from West to East, Link 1 (L01) is the section of Lower Stoke Road that joins Warminster Road. The colour of each link is the mean speed in both directions as per the scale shown in L06 from the Google Distance Matrix API queries. Below each section plot is the speed profile split by direction.

L02 is the section from Midford Brook bridge to the base of Winsley Hill.

L03 is Winsley Hill up to where the speed limit currently changes to 50 mph, at the junction with Woodlands Drive.

L04 is the 50mph section of the bypass, stretching all the way to the roundabout with Winsley and Bradford Roads. L05 is the 40 mph section of Winsley Road L06 is the 30mph section of Winsley Road.

Each segment was queried for each hour of a future week, which was the week beginning 5th January 2026, in both directions for each of the 6 segments. This was 168 x 2 x 6 = 2,016 calls in total. Google provides 5,000 free calls per month, with further calls charged at 0.5 pence per call.

Scenarios

The current speed limits along the road are shown in the figure below.

Given the complex nature of the road, with steep gradients, narrow, winding sections and proximity to residential areas there are numerous proposals. These have been broken down into scenarios

Scenario A

Reducing the 50mph section to 40mph bringing it into line with adjoining roads

Scenario B

Reducing the speed limit along the entire stretch of road to 30mph, bringing it into line with the Eastern most segment running through Bradford on Avon.

Scenario C

30mph along the full stretch but 20mph in the most built up section through Bradford on Avon, bringing it into line with other urban areas in Wiltshire.

Scenario D

20 mph along the full stretch, the most effective at reducing noise and collision risk.

Vision Zero

A 20 mph limit may appear extreme at first glance; however, Wiltshire Council has formally committed to the principles of Vision Zero, which aims to eliminate all deaths and serious injuries on the road network. At present, the county remains a considerable distance from this ambition, with 189 people Killed or Seriously injured on the regions roads in 2024. This suggests that more decisive and ambitious measures may be required.

Vision Zero is founded on the principle that if road users comply with the rules, the transport system should be designed so that fatal outcomes are not possible. Current vehicle safety standards (Euro NCAP crash tests) require vehicles to withstand a 64 kph (40 mph) impact with a rigid barrier. Any collision occurring at higher speeds exceeds this tested safety threshold.

On rural roads, head-on collisions are a common and particularly severe crash type (as evidenced in STATS19 data). In such cases, the equivalent impact speed is the combined speed of both vehicles. Two vehicles colliding head-on at 20 mph each produces an impact broadly equivalent to a single vehicle striking a solid object at 40 mph. Speeds above this therefore significantly increase the likelihood of fatal injury, even in modern vehicles.

Impact on journey time

Using the data from the Google API the impact on journey time along the stretch of road was assessed. This was done by taking the average speed for each hour, the length of each stretch of road and swapping the average speed for the revised speed limit. If the speed limit was lower than the average speed the journey time was recalculated with the speed limit assumed to be the speed.

The results of this for each scenario are shown in the table below for each direction

The current situation is not visible on the plot, because it is behind Scenario As line as it is identical (499 seconds Westbound and 490 Eastbound). Scenario B is estimated to have a 30 second increase in both directions, C 45 second increase Westbound and 40 seconds Eastbound and D 197 seconds increase Westbound and 191 Eastbound.

Economic impact

The DfT uses TAG data to attribute value to decisions. There has been a lot of criticism of this approach, especially when it is

The market price of journey times is often mentioned. But it also includes: - Monetary valuation of changes in noise - Damage cost and marginal abatement cost values by pollutant - £ per tonne of CO₂e emissions - Value of prevention per collision - Value of improving active travel infrastructure - Social impacts of public transport

Trip data

The DfT last estimated the trips along the B3108 at Winsley Hill in 2019 (https://roadtraffic.dft.gov.uk/count-points/947637). Prior to this a manual count was undertaken each year for the last 10 years. The manual count data in 2018 provides a diurnal profile for each vehicle and direction. The directional data estimated in 2019 has been combined with the diurnal profile.

In 2024 it was estimated that Annual Average Daily Trips (AADT) were similar to 2019. Therefore, 2019 traffic data has been used for this analysis.

On average 52% of vehicle traffic travels Eastbound. This reaches its peak at 6am when 74% of vehicles are travelling in this direction. At 5pm the majority (63%) of traffic flows in the Westbound direction.

Value of time

Tables A1.3.1-A1.3.5 from TAG have been matched to trip data for the road, disaggregating each trip by estimated economic activity. This has been combined with the journey time estimates to provide a figure for the perceived loss to the economy

Calculating for each scenario by direction gives:

The TAG perceived loss to the market for each Scenario over the course of a year is:
- A: £1,644
- B: £661,576
- C: £801,843
- D: £4,087,308

Evidence that uniform speed limits are less cognitively challenging

20mph is most efficient speed for electric vehicles, saving drivers money.

population

It is estimated over 1400 people live within 100 metres of the B3108. 225 of these live within 100 metres of the 50mph section.

Further details on how this was estimated is given in the appendix.

Collision costs

Between 2010 and 2024 there were 57 collisions resulting in 74 casualties reported to the Police, these are plotted below: In Bradford on Avon it is mostly Pedestrians and along the bypass it is predominantly motorists and Winsley Hill Cyclists.

Applying the methodology from TAG to the collisions along B3108 and summing per link gives an estimate for the value of prevention per year. ## Cycle infrastructure benefits Narrowing the road could be achieved by adding high quality cycle and walking infrastructure to the road, such as the AI generated image below. TAG table 4.1.6 provides a cost benefit for cycle infrastructure to the cyclist based on the time spent cycling the infrastructure.

Using the DfT estimate for cyclists using the road each day, it is estimated that the benefit for adding a segregated cycle path along the full length of the B3108, assuming a cyclist travels at 5 metres per second, could be quantified at over £30,000 per year to the 500+ cyclists who currently use the road every week. Adding this, along with a high quality footpath, could reduce the road width, also reduce the speed of vehicles down the road. It would also become an artery for active travel through the area.

One of the takeaways from the analysis of STATS19 data on drivers involved in collisions on the road is that they generally live locally. Enabling people to undertake local journeys on foot or by bicycle will reduce the number of motor vehicles on the road.

(map of LSOA of drivers)

Air pollution

The image below shows an example of NO₂ drop off calculated with the Defra NO₂ drop off tool. NO₂ reacts with the atmosphere and drops off relatively quickly from the roadside. Even without any changes to the vehicle fleet or number of vehicles using the road, it is estimated that increasing the source (vehicles) and receptor (buildings) distance by 2m NO₂ would be 14% lower at the receptors

TAG sheet A3.3 Table A 3.2.2 Gives a damage cost value range based on incremental concentrations. Assuming a roadside NO₂ concentration between 10ug/m3 and 40ug/m³ and taking the lower and upper monetary estimates the benefit of reducing roadside concentrations by 14% for the population living next to the road would be between £1,546 and £239,733 just for NO₂.

Summary

• Adding cycle path: £30,000
  
• Collision Avoidance: £245,205
  
• Reducing speed limit:
• Scenario A: -£1,643
  
• Scenario B: -£661,576
  
• Scenario C: -£801,843
  
• Scenario D: -£4,087,308
  
• Noise reduction:
  
• Air Pollution: £1,546 - £239,733

Appendix

Journey time

The table below lists the estimated journey time for each road link based on the Google API request

Value of time

The tables A1.3.4 and A1.3.5 from TAG data book were imported and reshaped to give a timeseries for each hour of the day.

Table A1.3.4

There are two methods for calculating time, per vehicle or per person. TAG table A1.3.3 gives estimates for occupancies of vehicles for time ranges during the week and weekend averages. However, this can become complicated when deciding trip reason for car occupants. Table A1.3.4 is shown below, reshaped into a value for weekday or weekend for each hour for each vehicle type.

Table A1.3.5 is shown below, reshaped into a value for weekday or weekend for each hour for each vehicle type.

These tables are joined to the DfT AADT data to give one table with flows split by trip purpose and assigned a cost.

Local population calculation

The Office for National Statistics (ONS) compiles regularly updated population data for the ‘Local Super Output Areas’ (LSOA). The plot below shows the total population for LSOA areas that intersect the B3108. In order to estimate the population close to the road the population was distributed over the OSM buildings shapes based on the area of the buildings. The population of each building that falls within a 100m buffer of either side of the road was summed. It is estimated that just over 1400 people live within 100 metres of the B3108, the majority along Winsley Road in Bradford on Avon.